Early intervention has increasingly become the new standard of care for many patients with breast cancer, with an ever-growing collection of phase III trials currently exploring several novel or improved approaches.
Gunter von Minckwitz, MD, PhD
Early intervention has increasingly become the new standard of care for many patients with breast cancer, with an ever-growing collection of phase III trials currently exploring several novel or improved approaches, according to leading researchers who spoke at the 33rd Annual Miami Breast Cancer Conference® (MBCC).
The paradigm shift was kick-started by the FDA’s approval of neoadjuvant pertuzumab (Perjeta) in September 2013. This approval was based on pathological complete response (pCR) rates from the phase II NeoSphere trial. In a 5-year follow-up analysis of the trial,1 patients who were treated with pertuzumab, trastuzumab (Herceptin), and docetaxel had better long-term outcomes than those who received trastuzumab plus chemotherapy, and those outcomes correlated with pCR rates.
AC indicates doxorubicin and cyclophosphamide; EC, epirubicin and cyclophosphamide; EFS, event-free survival; FEC, fluorouracil, epirubicin, and cyclophosphamide; gBRCA, germline BRCA; HR, hormone receptor; HER2, human epidermal growth factor receptor-2; iDFS, invasive disease-free survival; pCR, pathological complete response; TNBC, triple-negative breast cancer.
Importance of Pathological Complete Response
Pertuzumab was the first drug ever approved based on pCR as a surrogate endpoint, opening the floodgates to other novel agents, explained Gunter von Minckwitz, MD, during a presentation at MBCC, which Physicians’ Education Research (PER®) hosted in Miami Beach in March. Von Minckwitz is managing director of the German Breast Group and an associate professor and senior physician at University Women’s Hospital in Frankfurt. Now, there are a number of clinical trials evaluating novel regimens in the neoadjuvant setting as well as studies testing new strategies for the immediate aftermath of neoadjuvant therapy and in more traditional adjuvant scenarios. (Table).The use of pCR as a clinical trial endpoint helped lay the groundwork for earlier intervention. “With the first example of pertuzumab, we can speculate that patients with early breast cancer got access to this innovative compound probably 3 to 4 years earlier compared with the usual pathway,” von Minckwitz said.
Although it can be used to expedite approvals, pCR is not meant to support a full FDA indication. Instead, pCR represents an intermediate endpoint, which is a surrogate marker of treatment efficacy assessed earlier than the true outcome of interest, such as survival, von Minckwitz noted.
There are benefits associated with the use of pCR as an endpoint for drug approvals, specifically earlier access to promising compounds with new mechanisms of action, said von Minckwitz. However, there are risks associated with its use, such as a lack of information on long-term efficacy and toxicity.
In the case of pertuzumab, the accelerated approval was not based solely on data from the neoadjuvant setting, von Minckwitz noted. “We had extraordinary efficacy in the metastatic setting, it has a very favorable toxicity profile, and it has no detrimental impact on the relative dose intensity of the backbone treatment,” he said.
In the metastatic setting, the phase III CLEOPATRA study showed an unprecedented 15.7-month improvement in overall survival (OS) with the combination of pertuzumab, trastuzumab, and docetaxel compared with trastuzumab and docetaxel alone.2 In the pertuzumab arm, the median OS was 56.5 months versus 40.8 months in the control arm (HR, 0.68; 95% CI, 0.56-0.84; P <.001).
With neoadjuvant approvals, the FDA has mandated that findings should be validated in a second study, preferably conducted in the adjuvant setting.
To confirm the pertuzumab approval, the phase III APHINITY trial is currently exploring pertuzumab, trastuzumab, and chemotherapy as an adjuvant therapy for women with HER2-positive breast cancer.
This trial has fully accrued 4805 patients for a year of treatment with the triplet compared with the trastuzumab and chemotherapy alone. The estimated study completion date is December 2023.3
HER2-Positive Early Stage Breast Cancer
“The APHINITY trial, the adjuvant trial, is fully recruited, so it fits nicely with the two-trial approach that was proposed by the FDA,” said von Minckwitz. “The surrogacy of pCR is best shown for trastuzumab, and therefore there is the belief that pertuzumab behaves quite similar, so the risk—benefit ratio is quite favorable.”In addition to pertuzumab, neoadjuvant therapies for patients with HER2-positive breast cancer are being assessed in other clinical trials.
To date, the relationship between pCR and the end result of survival has been better established for patients with ER-negative/ HER2-positive disease versus those with ER-positive/HER2-positive breast cancer, according to Sunil Verma, MD, who presented on neoadjuvant approaches in HER2-positive breast cancer at MBCC.
“The ER-positive/HER2-positive patients do not have the same type of relationship, probably because many of them will end up receiving antiestrogen therapy behind their current neoadjuvant therapy, and it is hard to establish a relationship by just looking at pCR,” said Verma, who is medical director of the Tom Baker Cancer Centre and head of the Department of Oncology at the University of Calgary in Alberta, Canada. “However, I think that model and the relationship between pCR and OS is quite strong.”
Verma highlighted the antibody—drug conjugate T-DM1 (ado-trastuzumab emtansine; Kadcyla) as another neoadjuvant option on the horizon, specifically in combination with endocrine therapy. In the phase II ADAPT study,4 which included 376 patients with HER2-positive/ hormone receptor (HR)-positive breast cancer, patients received neoadjuvant T-DM1 with or without endocrine therapy or trastuzumab plus endocrine therapy.
The pCR rate was found to be substantially higher in the T-DM1 arms compared with trastuzumab plus endocrine therapy (P <.001). In the T-DM1—alone arm (n = 37), the pCR rate was 40.5%. In the arm with T-DM1 plus endocrine therapy (n = 48), the pCR rate was 45.8%. In the arm with trastuzumab plus endocrine therapy (n = 45), the pCR rate was 6.7%.
The phase III KRISTINE trial, which has fully accrued 444 patients, is currently comparing T-DM1 plus pertuzumab with chemotherapy plus trastuzumab and pertuzumab, the current neoadjuvant standard of care for HER2-positive breast cancer. The primary endpoint of the study is pCR.5
“Data from the ADAPT trial, which was conducted in Germany, showed that HR-positive and HER2-positive patients, when given T-DM1 for just 4 cycles, had a pCR rate of about 40%.
This is quite a significant level of activity seen for this subset of patients,” said Verma. “There are a number of other clinical trials, which have completed accrual and we will get the results of this year, including the KRISTINE study.”
In an interview with OncLive, Verma advocated for continued expansion in the use of neoadjuvant therapy, noting that treatment could be tailored based on whether a pCR is achieved to produce the best outcomes. Neoadjuvant therapy followed by surgery also allows for the tailoring of adjuvant approaches based on the level of residual disease that remains.
“The rationale behind neoadjuvant therapy is that we have quicker access to systemic therapy, and we have an opportunity to establish whether patients achieve pCR or not,” he said. “If you are going to give treatment to a patient with HER2-positive breast cancer and their tumors are palpable, they should receive neoadjuvant therapy.”
Neoadjuvant Therapies for TNBC
Studies are currently assessing continued therapy for patients with residual disease following neoadjuvant therapy and surgery, notably the phase III KATHERINE study, which builds upon the ADAPT trial.6 In this two-arm study, T-DM1 is being compared with trastuzumab as an adjuvant therapy in patients with residual disease in the breast or axillary lymph nodes following preoperative therapy and surgery. In the study, additional radiotherapy and/or hormonal therapy will be administered if appropriate.At this time, the only proven success for pCR, based on an FDA approval, has been for patients with HER2-positive breast cancer; however, there are high hopes for this endpoint in triple- negative breast cancer (TNBC), according to von Minckwitz.
In an attempt to further prove the connection between pCR and long-term outcomes, researchers conducted an analysis of 12 trials that explored neoadjuvant therapies for women with breast cancer.7 Overall, data from 11,955 patients were explored in this pooled analysis, which failed to established a correlation between pCR rates and event-free survival (EFS) and OS across the full population.
“One of the hypotheses to explain this is that it might be that these were not the ideal trials. They used chemotherapy agents, and sometimes not in the best way,” said von Minckwitz, who was a senior author on the study. “The pCR rates were sometimes not that large. Maybe these trials were not optimal for doing this analysis.”
Despite a lack of connection between increases in frequency of pCR and outcomes, the study did show a clear association between OS and pCR in the breast and lymph nodes versus the breast alone (HR, 0.51). This correlation was most prominent in patients with TNBC (HR, 0.16) and for those with HER2-positive/HR-negative breast cancer who received trastuzumab (HR, 0.08).
In TNBC, a number of ongoing studies continue to assess carboplatin as a neoadjuvant treatment, specifically in those with BRCA mutations.
In the phase III TNT study, carboplatin was compared with docetaxel in patients with metastatic TNBC. Additionally in the neoadjuvant setting, carboplatin was explored in the phase II CALGB 40603 and GeparSixto trials.
In the TNT trial, 376 patients with TNBC or germline BRCA1/2 mutated metastatic breast cancer were randomized to first-line treatment with docetaxel or carboplatin. Across the full study, the median progression-free survival (PFS) for carboplatin was 3.1 months versus 4.5 months for docetaxel.8 By this measure, “there was no clinically meaningful difference between the two treatments,” von Minckwitz concluded.
Where carboplatin stood out was in BRCA-mutated tumors. In the group with germline BRCA1/2 mutations (n = 43), the objective response rate with carboplatin was 68.0% compared with 33.3% for docetaxel (P = .03). In this same population, carboplatin showed a 6.8-month PFS compared with 4.8 months with docetaxel.
“This shows that carboplatin works much better in the BRCA-mutant tumors. It would be inferior for the nonmutant, and it looks to be superior in the mutant tumor compared to the docetaxel,” von Minckwitz said.
In the CALGB 40603 trial, 443 patients with stage II/III TNBC received neoadjuvant therapy with weekly paclitaxel followed by doxorubicin plus cyclophosphamide.9 Patients could also receive concurrent carboplatin plus bevacizumab or either drug alone.
After a median 39-month follow-up, carboplatin improved the pCR rate in the breast from 44% to 60% (P = .0018), while the addition of bevacizumab increased the pCR rate from 48% to 59% (P = .0089). In the breast/axilla, carboplatin significantly increased the pCR rate from 41% to 54% (P = .0029); however, the increase in pCR rate with bevacizumab of 44% to 52% was not statistically significant (P = .0570). Neither EFS nor OS were statistically improved with carboplatin and bevacizumab.
“They showed significant improvement in pCR, but they could not show a significant improvement in event-free survival,” said von Minckwitz.
“There was a trend but this was not significant. The trial was of course like a typical neoadjuvant study—not large—so the statistical power was not very high to show such a difference.”
In the phase II GeparSixto trial, 595 patients received weekly paclitaxel plus non-pegylated liposomal doxorubicin (Myocet) with (n = 296) or without (n = 299) carboplatin.10 Overall, the addition of carboplatin improved the pCR rate from 36.9% to 53.2% (P = .005). At 3 years, DFS was 85.8% with carboplatin versus 76.1% for those who received no carboplatin.
“In the GeparSixto study, we used carboplatin in a weekly quite dense way in combination with paclitaxel and Myocet, so already the baseline treatment was very intense,” said von Minckwitz, who was the lead author of the study. “And we could show still that in triple-negative patients, carboplatin increased the pathologic complete response by 20%, so a massive effect. And this was translated also in an improved disease- free survival in the smaller trial. So quite an intensive effect there as well.”
Contrary to the TNT study, the BRCA results were less telling in GeparSixto. In the study, 50 patients in the study had known germline BRCA mutations. In the BRCA wild-type population, the odds ratio (OR) for pCR was 2.09 in favor of carboplatin versus the control arm (50.8% vs 33.1%; P = .005). In the BRCA-mutant group, the OR for pCR was 1.6 in favor of carboplatin, but the difference was not statistically significant (61.5% vs 50.0%; P = .413).
Utilizing pCR for Future Approvals
“In the more sensitive BRCA-mutant tumors, just one hit with the anthracycline is already enough to take most of the advantage of DNA-damaging agent, so that the additional use of carboplatin does not have an additional effect in this group,” von Minckwitz hypothesized regarding the lack of benefit. “Whereas, in the nonmutant tumors, which might be maybe a little bit less sensitive to DNA-damaging agents, a second hit might be more important.”The exploration of pCR has been fraught with ups and downs, specifically as it relates to long-term outcomes; however, trends are beginning to emerge regarding the types of agents that are successful and indicators for potential success, such as gene mutations and proliferative index.
In the phase III NOAH trial, pCR was associated with EFS at 5 years in patients who received chemotherapy plus trastuzumab versus chemotherapy alone (HR, 0.29).11 A pCR was achieved by 45 patients in the trastuzumab arm versus 23 in the chemotherapy-alone arm. The 5-year EFS rates were 58% and 43%, with and without trastuzumab, respectively (HR, 0.64; P = .016).
“This fits with the large adjuvant studies that were published later on,” said von Minckwitz. “Here, with an antibody treatment, it seems to work quite well, and therefore there was a feeling that maybe pCR surrogacy is much better present when we use antibody treatments and targeted agents.”
However, this was not the case between the NeoALTTO and ALTTO studies, which explored lapatinib and trastuzumab. These large studies both failed to show a connection between improvements in pCR and survival but did further confirm that patients who achieved a pCR had better long-term outcomes.
“NeoALTTO was probably overoptimistic, several other neoadjuvant lapatinib studies are negative,” said von Minckwitz. “On the other hand, ALTTO was probably underpowered. It was a time-driven analysis performed with only twothirds of the required events.”
As it stands now, it is unclear whether pCR will still be instrumental in the approval of other neoadjuvant therapies for breast cancer. The next potential milestone decision could be based on the phase III BRIGHTNESS trial, which is exploring the PARP inhibitor veliparib (ABT-888) as a neoadjuvant therapy for TNBC, believes von Minckwitz.12
The three-arm BRIGHTNESS trial is evaluatung veliparib, carboplatin, and paclitaxel with placebo plus carboplatin and paclitaxel, or paclitaxel alone. The primary endpoint is pCR, with secondary endpoints focused on eligibility for breast conservation after therapy, survival, and quality of life. The study plans to enroll 624 patients with an estimated completion date of February 2017.